RAVEN

[Barthle, Jonathan]

Good afternoon,

I am currently working on a RAVEN file for a sensitivity analysis of a RELAP5-3D model. My intent is to add samples to the Monte Carlo sampler after each batch. I have reviewed the manual and user guide to try and implement a restart to save the previous runs. However, I am currently running into an issue where my previous runs are being wiped. For example, if I had 100 runs initially and I want to add 100 more, the original 100 runs are wiped and RAVEN samples 200 new samples instead. I was wondering if I could have some guidance as to what I might be doing wrong. Attached are some of the relevant nodes in the RAVEN file. Let me know if anything else is needed from my end.

Thank you

Jonathan

[Andrea Alfonsi]

Hi Jonathan,

can you send the complete input file (XML)? Redact all the variables that you don't want to show.

Thanks a lot.

Andrea 

--
You received this message because you are subscribed to the Google Groups "INL RAVEN Users Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to inl-raven-use...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/inl-raven-users/CAMsK15KK9-pvGXdY5XEH723JuDoy6Wz7_0QNJ5YvOG2CxJuvwQ%40mail.gmail.com.

[Barthle, Jonathan]

Sure thing, attached is a copy of the xml file without all of the variables.

Thank you

Jonathan

[Congjian Wang]

Hi Jonathan,

Can you try to add additional node to provide restart tolerance? Our default restart tolerance is 1E-14. Since you restarted from a CSV file, I guess you may have different numerical tolerance there. Here is an example:

Best,

Congjian 

---

From: 'Barthle, Jonathan' via INL RAVEN Users Group <inl-rav...@googlegroups.com>
Sent: Monday, April 14, 2025 9:55 AM
To: andrea.a...@gmail.com <andrea.a...@gmail.com>
Cc: inl-rav...@googlegroups.com <inl-rav...@googlegroups.com>
Subject: Re: RAVEN

 

To view this discussion visit https://groups.google.com/d/msgid/inl-raven-users/CAMsK15KTeOV66rGJFo2bAaBfSq0YzCgNz8ggvMR_wU0OSbZnQQ%40mail.gmail.com.

[Barthle, Jonathan]

Good afternoon,

I attempted to add the restart tolerance node today. However, I am running into trouble with running RELAP5-3D. I am attempting to run it on Bitterroot and am getting this error message:

prterun was unable to find the specified executable file, and therefore did
not launch the job.  This error was first reported for process rank
0; it may have occurred for other processes as well.

NOTE: A common cause for this error is misspelling a prterun command
   line parameter option (remember that prterun interprets the first
   unrecognized command line token as the executable).

Node:       br033 Executable: relap5

Thank you
Jonathan

[Congjian Wang]

Hi Jonathan, 

I guess you need to do module load relap53D. You can find the RemoteRunCommand file raven_ec_slurm_command.sh under raven/ravenframework/, you can add "module load relap53D" in raven_ec_slurm_command.sh right after module load openmpi. 

Best,

Congjian 

---

From: 'Barthle, Jonathan' via INL RAVEN Users Group <inl-rav...@googlegroups.com>
Sent: Monday, April 14, 2025 9:55 AM
To: andrea.a...@gmail.com <andrea.a...@gmail.com>
Cc: inl-rav...@googlegroups.com <inl-rav...@googlegroups.com>
Subject: Re: RAVEN

 

To view this discussion visit https://groups.google.com/d/msgid/inl-raven-users/CAMsK15KTeOV66rGJFo2bAaBfSq0YzCgNz8ggvMR_wU0OSbZnQQ%40mail.gmail.com.

[Barthle, Jonathan]

This is what I currently have in my raven_ec_slurm_command.sh. What is strange is that I have been running RELAP normally up until today. Today is the first day I have not been able to run RELAP in RAVEN. 

module load openmpi
module load use.exp_ctl
module load relap53D/relap53D-execute

Thank you

Jonathan

[Congjian Wang]

I guess you may need to talk to HPC. Please keep us updated. 

Congjian 

---

From: Barthle, Jonathan <jbar...@vols.utk.edu>
Sent: Tuesday, April 15, 2025 12:10 PM
To: Congjian Wang <Congji...@inl.gov>
Cc: andrea.a...@gmail.com <andrea.a...@gmail.com>; inl-rav...@googlegroups.com <inl-rav...@googlegroups.com>
Subject: Re: RAVEN

 

[Diego Mandelli]

In the meantime, can you run the RELAP5 input file using RELAP5 directly (i.e., without RAVEN)?

 

To view this discussion visit https://groups.google.com/d/msgid/inl-raven-users/BY5PR09MB5490B131518E8F8D497593BAF9B22%40BY5PR09MB5490.namprd09.prod.outlook.com.

[Barthle, Jonathan]

It appears that this is an HPC/RELAP issue and not a RAVEN issue. I attempted to run RELAP by itself and ran into the same error. I have reached out to HPC support and will keep you updated in the meantime.

Thank you

Jonathan

[Diego Mandelli]

Okidoki, let us know about any new development.

Diego

[Barthle, Jonathan]

Good morning,

It turned out to be an issue on the HPC side and it was fixed. 

I have added the restart tolerance node to the xml file to see if it would resolve the problem I have been having with restarting. It still seems that the sampler is still resetting the already existing samples and starting over from zero when I try to restart. Could there be something wrong with my Steps or something missing from the Sampler?

Thank you 

Jonathan

[Diego Mandelli]

Jonathan,

Did you make additional changes to the input file you have been sharing?

[Barthle, Jonathan]

Good afternoon,

I have not made any additional changes relative to the file I shared. I should be using the exact same one.

Thank you

Jonathan

[Barthle, Jonathan]

Actually, I did add <restartTolerance>1e-9</restartTolerance> after the <restart> node to see if that would fix the problem.

Thank you

Jonathan

[Congjian Wang]

Hi Jonathan,

Could you share the raven screen output with us, as well as some examples from the raven output csv file? In addition, I may suggest to increase the restartTolerance.

Congjian 

---

From: Barthle, Jonathan <jbar...@vols.utk.edu>
Sent: Tuesday, April 22, 2025 12:54 PM
To: Diego Mandelli <diego.m...@inl.gov>
Cc: Congjian Wang <Congji...@inl.gov>; andrea.a...@gmail.com <andrea.a...@gmail.com>; inl-rav...@googlegroups.com <inl-rav...@googlegroups.com>
Subject: Re: RAVEN

 

[Barthle, Jonathan]

Sounds good, attached is the RAVEN output as well as the output csv file with the samples. I have included the original csv (HTGRTR_MC.csv) as well as the new version that is supposed to have the additional 100 samples, but instead just has 200 new samples (HTGRTR_MC2.csv). Let me know if there are any issues.

Thank you

Jonathan

[Congjian Wang]

It seems to me the initial seed for MC is not provided, and a random seed is used. You may want to try to add your own seed if you would like restart from existing file. The following is an example: 

Congjian 

---

From: Barthle, Jonathan <jbar...@vols.utk.edu>
Sent: Tuesday, April 22, 2025 3:03 PM
To: Congjian Wang <Congji...@inl.gov>
Cc: Diego Mandelli <diego.m...@inl.gov>; andrea.a...@gmail.com <andrea.a...@gmail.com>; inl-rav...@googlegroups.com <inl-rav...@googlegroups.com>
Subject: Re: RAVEN

 

[Barthle, Jonathan]

Sounds good, I will give it a try.

Thank you

Jonathan